This invention relates to improved pyroelectric film sensors.
Thermal sensors for infrared radiation are widely used in applications from motion detection to thermal imaging. Of the many available kinds, far and away the most economical are pyroelectric detectors made from a thin film of poled ferroelectric polymer, poly(vinylidene difluoride) (PVDF).
The more familiar ferromagnetic materials, when magnetized by an applied magnetic field, exhibit a remanent magnetization even when the external field is removed. Similarly, a ferroelectric is a dielectric that, after being poled, exhibits a remanent electric polarization in the absence of any applied voltage. Such a poled ferroelectric is called an electret. The remanent polarization is strongly temperature dependent, dropping to zero at a temperature characteristic of the material, its Curie point, which for PVDF is about 120 C. It is this temperature dependence that makes them useful for temperature measurements.
Just as a ferromagnet""s field can be described as arising from bound currents circulating in the ends of its pole pieces, so a ferroelectric""s polarization can be described usefully as a bound charge, located in a sheet just inside the polymer layer.
Because this bound charge will produce a nonzero electric field outside the film, it attracts free charge to the surfaces until overall electrical neutrality is achieved. As the film""s temperature changes, the amount of surface charge required also changes, and if these charges are made to flow through an external circuit, they form the basis of a temperature measurement scheme of great sensitivity, enough to be a competitive thermal radiation detector for objects near room temperature.
Our work includes a summary critique of important prior art that refers to pyroelectric film sensors, as well as review of a sensor""s underlying physics, to an end of developing new insights and methodology that can be of value in disclosing a new and improved pyroelectric film sensor.
One major drawback of film sensors as optical sensors is the difficulty of coupling them to the radiation field; they reflect most of the incident infrared radiation from their metallized surfaces. The figure of merit for an emitter or absorber of thermal radiation is its emissivity, which is the ratio of the total power radiated or absorbed per unit area to that of a perfect black body in similar conditions. The emissivity is always between 0 and 1, and is always equal to the absorptivity, which is the corresponding figure of merit for absorption.
In order to capture the displaced charge, the surfaces of the film must be coated with an electrical conductor; though the currents are typically very small, so that no great thickness is required, still the film must be thick enough to survive handling and to maintain electrical continuity between all areas of the sensor element and the external wiring.
Although 25 nm films of certain metals such as Inconel, nickel, and rhodium make respectable absorbers in the visible, in the mid and far IR such films look like perfect conductors, keeping the infrared emissivity down near 0.02 and the measurement""s sensitivity and signal to noise ratio (SNR) accordingly 34 dB below their theoretical maxima, as shown in FIG. 1 (numeral 10). In the visible, it is not difficult to make a surface a better absorber: just paint it black. A black surface is one which attenuates light gradually over a depth of many wavelengths; a lossier material (i.e., a better conductor), whose absorption depth is less than a wavelength starts to look like a metal instead. In the infrared, metals become essentially perfect conductors, so the absorption depth becomes small just as the wavelength increases, making it dramatically more difficult to build efficient absorbers that are not extremely thick.
Because usually one wants to build pyroelectric sensors that are as fast as possible, thick coatings that would slow down thermal diffusion and add a huge thermal mass are inadmissible, which is a problem. It really isn""t a matter of the surface properties of the film, because in order for any surface to be really black, it has to be many wavelengths deep. At 10 microns, that means a big thick structure that will have an enormous thermal mass, which will dominate the response time of the sensor.
In the prior art, there is a significant amount of discussion of thin-film solutions to this problem, using thin metal films on the top and bottom surface of a free-standing polymer film.
One especially relevant paper is S. Bauer, S. Bauer-Gogonea, W. Becker, R. Fettig, B. Ploss, W. Ruppel, and W. von Munch, xe2x80x9cThin metal films as absorbers for infrared sensorsxe2x80x9d, Sensors and Actuators A, v37-38, pp. 497-501 (1993).
The authors"" main point is that optically thin metal films (less than one absorption length) look like the optical equivalent of a lumped-constant RLC network in RF circuits, and that a pyroelectric film sensor can be given extremely good broadband absorption properties (e.g. absorptivity  greater than 0.5 and completely flat from 3-200 microns), all by adjusting the film thickness on the front and back. The two metal layers, 188 ohms/square on the top of the film (where the radiation hits) and 800 ohms/square on the bottom, reach the optimal absorption with almost no impact on the thermal mass whatever (e.g., see FIG. 2, numeral 12).
These films are extremely thin, just a few nanometers, and correspondingly fragile. Since as mentioned above, the surface films must function as wiring as well as absorber, this is a serious drawback. It is very difficult to make a nanometer-thickness coating be continuous across a centimeter-size piece of pyroelectric film, and even more difficult to keep it continuous under handling (e.g., see FIG. 2). Thus there is apparently a dilemma: one can improve the sensitivity of these sensors by 25 times (28 dB) by thinning down the metal coatings, but one can""t connect to them.
One method for solving this problem is to use two lithography steps on each side of the film. A film with the thin absorber layers can be patterned into pixels in one mask step, and then wiring can be deposited in a second step, either by screen-printing conductive ink or by evaporation, sputtering, or plating through a photoresist layer that is subsequently removed, much as is done in semiconductor processing. The problem is that multistep lithography on free-standing films is difficult and expensive, whereas the simple single-layer films can be patterned almost trivially using printed-circuit technology.
The present invention, in sharp contrast to the above discussion referencing prior art techniques and physics, as it pertains to pyroelectric film sensors, proceeds in a qualitatively different methodological manner. In particular, an object of the present invention is to exploit the high conductivity of the thick metal and the high absorptivity of the thin metal system in combination, to reap most of the sensitivity improvement while keeping the reliability and simplicity of the single thick metal layer approach.
Accordingly, in a first aspect of the present invention, a method for constructing a thin-film infrared photodetector includes providing a thin film of ferroelectric polymer, and applying at least one electrically conductive coating to the polymer so that the coating is nonuniform in surface conductivity for providing an area of higher surface resistance connected to an output region by an area of lower surface resistance. The method may advantageously further include patterning the film so as to change the electrical connectivity of an area.
In the invention, in a second aspect, a method for improving the performance and manufacturability of thin-film infrared photodetectors includes depositing a very thin layer of an absorbing material including a metal on a surface of detector material, and depositing a pattern of a thicker layer including electrically conductive material which is electrically continuous over at least a portion of the thin layer but contains spaces which allow incident radiation to reach the thin layer.
In a third aspect of the present invention, a photodetector includes a thin-film including ferroelectric polymer, the thin-film including at least one electrically conductive coating characterized in that the coating is nonuniform in surface conductivity, for thereby providing an area of higher surface resistance connected to an output region by an area of lower surface resistance.